Anticipating the New Generation of Digital Threats
Anticipating the New Generation of Digital Threats
Publish Date: 2026-05-27 03:39:00
Source Domain: www.telefonica.com
In less than five minutes, it is now possible to clone a person’s face and voice using Artificial Intelligence.
This idea, which just a few years ago seemed like something out of science fiction, was one of the central topics at the cybersecurity training day recently held by the Civil Guard in Valencia. There, officers from Team @ —specialised in cybercrime— worked on a scenario that is already part of the present: a new generation of AI-driven threats.
The session featured the participation of Telefónica, which contributed expert knowledge on how these technologies are transforming both cybercrime and the strategies used to combat it.
A Change of Pace in Digital Threats
One of the main challenges that Artificial Intelligence introduces to cybersecurity is acceleration.
Changes no longer occur over months, but in a matter of days. Tools evolve rapidly and, with them, so do attack techniques. This forces a rethinking of traditional defence models, which have historically been more reactive.
Furthermore, AI has significantly reduced barriers to entry. Activities that previously required advanced knowledge or substantial technical resources can now be carried out by a single person from a computer. This multiplies the potential scale of the attacks.
Deepfakes: The Challenge of Identity
Within this new context, deepfakes have become one of the main sources of concern.
The ability to generate synthetic content — voice, image, or video — with a high degree of realism introduces a new level of complexity. Identity theft is no longer a rudimentary attempt, but rather a simulation that is difficult to detect, even for experienced users.
This scenario has direct implications in areas such as customer relations, identity verification, and fraud protection.
Anticipation as a Strategy
In light of these developments, cybersecurity needs to move towards more proactive models.
Artificial Intelligence itself, when applied correctly, can improve threat detection,…
